ML-algorithms-from-scratch-using-python

Dataset

Attributes of the dataset and their description

Algorithms Implemented:

(1) Linear Regression ( simple + multiple )
(2) Decision Tree ID3
(3) K-means clustering
(4) Logistic Regression
(5) Naïve Bayes
(6) Support Vector Machine (SVM)
(7) K-Nearest Neighbours (KNN)
(8) Principal Component Analysis (PCA)

Other things that are also implemented:

(1) EDA ( Exploratory Data Analysis )
(2) Normalization techniques: Min-Max, Z-score, Decimal Scaling
(3) Numeric similarity measures: Euclidean , Manhattan , Supremum , Mahalanobis

TODO:

Random forest

Gradient boosting algorithms

Various other regression algorithms like: Ridge, Lasso

Artificial Neural network

NOTE:

This particular dataset was chosen as it contains various data types(string, integer, float) and a very large number of records( approx 19,000 ) and attributes (72)

Short-comings of this dataset: Highly biased, unbalanced records.